No photos to EPUB from RTF files

[DVC]

After being told that PDF can't be used to get an EPUB -- after I posted that when I tried this conversion and ut produced terrible output -- I tried RTF.

Great conversion -- except all photos (which ARE in the RTF file -- are not moved to the EPUB file.

So PDFs can't be used! And, RTFs can't be used! Plus, .docx isn't input.

Why offer PDF and RTF as input formats if Calibre will not convert to EPUB?

[itimpi]

PDF can be used to get an ePub - it is just that any sort of complex formatting tends to mess things up. However novels for instance tend to convert fine. There is also the improved PDF conversion engine that is under development that will be released sometime in the near future (I guess after Kobid gets back from holiday) that is meant to be better, but we do not yet know how much.

As to RTF files they all convert fine for me (including pictures). However I suspect that it is very relevant exactly what format the pictures are in - in youir case they may be a file type that Calibre does not handle. When I have had problems with some RTF files they have often converted better if I load them into Microsoft Word, and then save them as Web page (filtered) and use the resulting HTML in Calibre.

[DoctorOhh]

Instead of expressing anger, how about trying to educate yourself on the processes involved in converting files. Ask for help with your situation and understand that someone here most likely has run into your situation and can help out.

Calibre is a labor of love by its creator, Kovid and he is amazingly patient in answering folks questions on here even if those folks choose to express themselves in a rude manner. He is currently out of the country on vacation and will be back around the third week of April, but he has been dropping in now and then with help.

Originally Posted by DVC

So PDFs can't be used! And, RTFs can't be used! Plus, .docx isn't input.

I've converted many a pdf and rtf file using Calibre.

Many PDFs though by nature are not easily converted to a flowable format that works with a 6inch ereader. Also if the doc you are using links to photos instead of those photos being embedded in the document there could be an issue.

I have always converted my problematic PDF and RTF files to HTML first just to take a look before converting them to epub. I use Mobipocket Creator (free) to convert PDF to HTML and Word to save RTF as filtered HTML.

Sometimes I use the Calibre debug feature to take a look at the doc as it is being converted. This feature is great and allows you to see the file at each point in the process and see what is happening at each step.

From Calibre's FAQ:

What are the best source formats to convert?

In order of decreasing preference: LIT, MOBI, EPUB, HTML, PRC, RTF, PDB, TXT, PDF

You can see that PDF is at the bottom of the list of preferred source formats to use for conversions. Kovid is in the middle of rewriting the PDF conversion process but it could be a while before it is released.

The user manual is very extensive, check it out for other tips.

Update: Once you have an epub file you can always touch it up with Sigil an epub editing package.

[kovidgoyal]

Most likely the image sin the RTF file are in an unsupported format. As itimpi said, use word to save the rtf as html.

And I don't apreciate your tone, no one is forcing you to use calibre, feel free to find an alternative that works better for you.

[Patricia]

What are the best source formats to convert?

In order of decreasing preference: LIT, MOBI, EPUB, HTML, PRC, RTF, PDB, TXT, PDF

If I were in the original poster's position, then I think that I'd be making a word doc, including all the pictures etc, then saving it as a lit book. After that I'd run it though Calibre.

You can get the Microsoft lit plug-in free here:
http://www.microsoft.com/downloads/d...DisplayLang=en

Here's our wiki entry on LIT:
https://wiki.mobileread.com/wiki/LIT

[JSWolf]

Originally Posted by itimpi

PDF can be used to get an ePub - it is just that any sort of complex formatting tends to mess things up. However novels for instance tend to convert fine.

Actually, they do not convert fine. A novel length text PDF will convert ok at best. It will have errors. No program can convert a novel length PDF to any other format without errors.

[Patricia]

That's very true, Jon. At the very least there will be an awkward line-break where the PDF starts a new page. Usually the italics disappear too.

And this happens whatever one uses for the conversion. I've tried Book Designer, Calibre and Mobipocket Creator. I just resign myself to the fact that there will never be a perfectly accurate automatic conversion, then set about editing it.

[HarryT]

Very often to best way to convert a PDF file to a more usable format is to use an OCR program rather than to try and convert the PDF directly.

[Patricia]

Originally Posted by DVC

RTFs can't be used! Plus, .docx isn't input.

It takes a couple of seconds to open a docx file then to use "save as" and save it as a doc.

I can see that you are frustrated, DVC, but there are often quite simple workarounds.

[HarryT]

Patricia knows as well as I do that there are no shortcuts to producing a really good-looking, thoroughly-proofed, eBook. It's a process that cannot be entirely automated, and which does involve a lot of "human" decision making. Proof-reading is an especially time-consuming - but absolutely vital - activity.

[animedude01]

Conversion is always difficult because there are so many variables. How was the original created? Is the PDF straight text or scanned images? Remember, PDF is a container and there can be almost anything inside.

If you have a .docx, use Word or OpenOffice to save it as an HTML file, edit the HTML to get as clean a source as possible, then use Calibre or Sigil to convert the HTML to ePub. You may still need to edit the ePub to tweak things.

[JSWolf]

Originally Posted by Patricia

That's very true, Jon. At the very least there will be an awkward line-break where the PDF starts a new page. Usually the italics disappear too.

And this happens whatever one uses for the conversion. I've tried Book Designer, Calibre and Mobipocket Creator. I just resign myself to the fact that there will never be a perfectly accurate automatic conversion, then set about editing it.

Even Adobe Acrobat Professional makes errors in the conversion.

[DVC]

Lot's of answers, but they all make my point.

First, .doc is not supported so that route is closed.

Second, converting to HTML will destroy a book's format. The result will be as bad as a web-page. I know, I've used Word to make HTMLs.

LIT? What the hell is that? I don't know where you live, but there is only one universal document standard and that is PDF. Since the rules of decoding PDF are decades old, why would anyone design a conversion program that accepted PDF as the "last" choice? In reality, PDF is the only input format needed for making eBooks.

Why list RTF if it can't convert an RTF with pictures. Telling me there may be pix format that is not supported only means RTF is not really fully supported. RTF means RTF as it comes from any software that produces it. Since almost every RTF document includes pix -- a conversion program has be able to deal with all pix. What's so radical about JPEGs?

And, the suggestions that one convert to X and then convert to Y to get to Z are just plain silly. At each step, more formating will be screwed-up. When one uses a graphics conversion program -- it just does the job.

And, since EPUB is designed for BOOKS -- the comments that one cannot expect to convert BOOKS kind of defeats the purpose of a CONVERTER TO EPUB. Now, were there a way to interactively edit the EPUB output so one could fix conversion problems -- it would be acceptable given the likelihood of conversion errors. But, there isn't. The Debug feature is a joke -- a folder full of .jpegs.)

Sorry, to be so negative, but Caliber simply doesn't make eBooks than can be sold. So, I'm not sure why it was written. Converting LIT to EPUB. Why?

[Timoleon]

"Sorry, to be so negative, but Caliber simply doesn't make eBooks than can be sold. So, I'm not sure why it was written. Converting LIT to EPUB. Why?"

Why not? LIT and EPUB are basically wrappers for HTML, and are related, so it makes perfect sense to me to convert one to the other.

Still don't know where you are coming from on all this...

[Elfwreck]

Originally Posted by DVC

Lot's of answers, but they all make my point.

First, .doc is not supported so that route is closed.

Anything that makes .doc & .docx files will save as .rtf instead. RTFs are often much larger files than .doc, but the formatting is all there. Supporting .doc may require licensing agreements with Microsoft, which are beyond the scope of most open-source projects. Or it might just involve an extra bit of code to convert .doc files to .rtf internally... in which case, since it is an open source project, anyone who's interested is welcome to write the code for that.

Second, converting to HTML will destroy a book's format. The result will be as bad as a web-page. I know, I've used Word to make HTMLs.

Word docs saved as HTML make lousy webpages, but often good ebooks. Ebooks work best as markup-based documents, like HTML and PDB (which I'm not recommending you deal with) not layout-based documents like PDF and word processor documents.

LIT? What the hell is that?

LIT is Microsoft's ebook format, one of the earliest commercial ebook types, and has some of the better complex formatting options. That you have no idea what it is indicates you're new to ebooks in general.

I don't know where you live, but there is only one universal document standard and that is PDF.

No, there's not. If there's a universal doc standard, it's txt.
PDF is new, rapidly-changing, and contains so many kinds of content that it's almost impossible to give anyone advice on how to work with PDFs. It's a very useful format--I love PDFs--but because it's got so many options about what can go in it, it's not always easy to get content out of it in a useful format.

However, you could bypass Calibre altogether by taking your RTFs and formatting them for the size of screen the ebooks will be read on. I make a lot of 6"-screen sized PDFs for my own reading. That way, I get exact control of the layout, the images, the fonts and so on.

Since the rules of decoding PDF are decades old, why would anyone design a conversion program that accepted PDF as the "last" choice? In reality, PDF is the only input format needed for making eBooks.

PDF *EN*coding is more than 15 years old, and has only been common for about the last 7 years. (Not "decades." Really. PDF was created in 1993.) *DE*coding is younger, and changes every couple of years with new editions of Acrobat, as PDFs gain more abilities. I've been working professionally with PDFs for 10 years, and it's changed a *lot* in that time. I still miss features from Acrobat 4.0, but I wouldn't want to go back to them & lose some of the options now available to PDFs.

Why list RTF if it can't convert an RTF with pictures. Telling me there may be pix format that is not supported only means RTF is not really fully supported. RTF means RTF as it comes from any software that produces it. Since almost every RTF document includes pix -- a conversion program has be able to deal with all pix. What's so radical about JPEGs?

PDB can only support .png images. If a converter didn't support moving images from RTF to PDB because they were JPGs, that doesn't mean RTF conversion isn't supported.

That said, ePub can support several image types. However, the formatting of the images in the RTF file may affect how (or whether) they convert to another filetype--inline images may convert, but images in frames may not. Images that are links instead of embedded may not import. Images of a type not displayable by the reader (browser, mobile device, e-ink screen, whatever) may be present in the file, but not viewable.

Getting a document to convert correctly involves knowing all the filetypes in the book, and how they intereract.

You are, of course, welcome to code your own ePub files by hand; they're HTML in a zip file, with a metadata file that basically tells the reader "this is an ePub file." That Calibre tries to assemble all the relevant data from all sorts of initial formatting is amazing.

And, since EPUB is designed for BOOKS -- the comments that one cannot expect to convert BOOKS kind of defeats the purpose of a CONVERTER TO EPUB. Now, were there a way to interactively edit the EPUB output so one could fix conversion problems -- it would be acceptable given the likelihood of conversion errors. But, there isn't. The Debug feature is a joke -- a folder full of .jpegs.)

There is a way to edit the ePub output. Change the .epub extension to .zip, unzip the folder, and edit away in your HTML editor of choice.

Sorry, to be so negative, but Caliber simply doesn't make eBooks than can be sold. So, I'm not sure why it was written. Converting LIT to EPUB. Why?

I don't make ebooks to sell them; I make them to read. If you're making ebooks to sell, it's a bit ridiculous to complain that a free, open-source program isn't making them the way you want them.

If you want to make sale-able ebooks with complex formatting, you may want to use a program like Bookdesigner that gives more precise control. Calibre can give a lot of control, but it's through extra lines of coding; Bookdesigner has a GUI for its editing. However, it's not simple & quick to learn.

Or you can learn to hand-code ePub files. I'm told that once the basics are learned, it can go fairly quickly. For myself, I make a lot of 3.46"x4.57" PDFs, because I'm comfortable tinkering with formatting in Word, and haven't figured out HTML/CSS editing yet.